hugetim

Contrary to what the leaderboard lists as the human score, their technical paper implies a human baseline of ~48%.

Am I crazy to think these models have actually surpassed human performance on ARC 2? https://www.lesswrong.com/posts/DX3EmhmwZjTYp9PBf/ai-perform...

The package Jeremy Howard uses for Claudette, nbdev, actually does address the notebook pain points of version control, testing, integration, and deployment: https://nbdev.fast.ai/ (Debugging is already a strong point of developing within a notebook.) But it's admittedly used mostly for solo projects or small teams and may not be ready for large orgs.

Curious how folks think this measures up: https://www.answer.ai/posts/2024-06-23-claudette-src.html

A nice source of middle- and high-school aged problems: https://www.usamts.org/

One of the joys of high school was discussing these with a friend (after submission---but one could forego the contest and just do the circle thing with them for fun).

For me it was this curriculum, which should be better-known: https://www.elementsofmathematics.com/

One place it is offered in-person is the St. Louis area where I grew up: https://megsss.org/about-us/

He's already written a bit here, starting in the second paragraph under "Why": https://about.fastht.ml/components

But is it reasonable to consider charlie worse than alice?

Clarify your thinking about what it means to ask "why" here: https://plato.stanford.edu/entries/questions/#PraAppExpCon

I'd love to see them implement a good liquid democracy system that would offer a better "middle ground between 'take every vote' and 'let Blackrock decide'."

I'm guessing you're not very familiar with Lessig's work: https://perma.cc/5XQ8-MBWA?type=image

Or maybe you just have a different theory of change? His seems to be that our best shot is to use whatever levers remain available in our nominal democracy to reduce the political power accruing to extreme wealth. What's yours?